Supervised Learing
- Anything you can do with 1 second of thought, we can probably now or soon automate.
Exmaple(email)
The toy arrived two days late, so I wasn't able to give it to my niece for her birthday.
Can I return it?
- Input text Refund / Shipping / Other
What happens if you try?
- 1000개의 데이터로 온전히 학습하기란 쉽지 않다.
- 물론 정확히 필요한 데이터의 수를 파악하는 것도 불가능하다.
What makes an ML problem easier
1. Learning a "simple" concept
2. Lots of data availabe
출처: Coursera, AI For Everyone, DeepLearning.AI
'AI For Everyone > 1주차' 카테고리의 다른 글
| Non-technical explanation of deep learning(Part 1,2 optional) (0) | 2022.09.22 |
|---|---|
| What makes an AI company? (0) | 2022.09.22 |
| Machine Learning / What is data? (2) | 2022.09.22 |
| Introduction (0) | 2022.09.21 |
